A mechanical fault monitoring method for optoelectronic composite submarine cable

Abstract

The present invention discloses a method for monitoring mechanical faults of an optoelectronic composite submarine cable, comprising: obtaining the voltage and current of the optoelectronic composite submarine cable, and calculating the loss value according to the voltage and current; obtaining the temperature data of the optoelectronic composite submarine cable; determining the fault evaluation index of the optoelectronic composite submarine cable based on a BP neural network; adjusting the monitoring frequency according to the fault evaluation index, and prejudging the safety status of the submarine cable. Through real-time detection and prejudgment of the status of the optoelectronic composite submarine cable, early judgment and timely warning of safety hazards are achieved, effectively ensuring the safe and stable operation of the submarine cable.

Description

Technical Field

[0001] The present invention relates to the technical field of submarine cable monitoring, and in particular to a method for monitoring mechanical faults of a photoelectric composite submarine cable. Background Art

[0002] With the development and utilization of marine resources, people are paying more and more attention to marine rights and interests, and submarine power cables have been increasingly widely used. Submarine cables are an important component of cross-sea interconnection projects and play an important role in the internationalization of power grids and the interconnection of regional power grids. With the development of the marine economy, the application of submarine cables in my country has also increased. Submarine cable communications are widely used in submarine scientific observations, military and coastal defense communications, marine resource development and other fields due to their advantages of long transmission distance, large capacity, low loss, high reliability, anti-interference and good confidentiality. Therefore, monitoring the operating status of submarine cables and performing fault diagnosis are the prerequisites for subsequent fault prevention and repair, and are very important for ensuring the safe and stable operation of submarine cables. With the frequent occurrence of submarine optical cable blocking accidents caused by human factors such as maritime transportation, fishing, and military activities, the economic losses are huge and seriously affect the security of my country's international communications. Therefore, it is necessary to propose a mechanical fault monitoring method for optoelectronic composite submarine cables to perform online monitoring and fault diagnosis of submarine cables. Summary of the Invention

[0003] The present invention collects the current, voltage and temperature data of the optoelectronic composite submarine cable, determines the safety status of the submarine cable based on the BP neural network, monitors the mechanical failure of the submarine cable, ensures the safe and stable operation of the submarine cable, and provides a data basis for the prevention and repair of mechanical failures of the submarine cable.

[0004] The technical solution of the present invention is:

[0005] A method for monitoring mechanical faults of an optoelectronic composite submarine cable, comprising:

[0006] Step 1: Obtain the voltage and current of the optical-electric composite submarine cable, and calculate the loss value based on the voltage and current;

[0007] Step 2: obtaining temperature data of the optoelectronic composite submarine cable;

[0008] Step 3: Determine the fault evaluation index of the optoelectronic composite submarine cable based on the BP neural network;

[0009] Step 4: Adjust the monitoring frequency according to the fault evaluation index and make a prediction on the safety status of the submarine cable.

[0010] Preferably, the loss value is calculated as:

[0011]

[0012] Among them, W loseIndicates the loss value, represents frequency, C represents capacitance, represents voltage, I represents the current carrying capacity of the submarine cable, G Σ represents the conductance value, θ represents the phase angle, and e represents the base of the natural logarithm.

[0013] Preferably, the temperature data is obtained based on finite element simulation analysis, which specifically includes the following steps:

[0014] Use ANSYS's adaptive meshing technology to perform triangular meshing on submarine cables;

[0015] Steady-state thermal analysis of a submarine cable immersed in seawater: Where h represents the convective heat transfer coefficient, T w Indicates the surface temperature, T f represents the fluid temperature, λ represents the simulation coefficient, and n represents the grid division number;

[0016] The loss value is applied to the corresponding loss area for simulation and solution to obtain the submarine cable temperature and conductor temperature.

[0017] Preferably, step three specifically includes:

[0018] Step a, establishing a three-layer BP neural network;

[0019] Step b: Determine the input layer neuron vector x = (x1, x2, x3, x4, x5) T , where x1 represents the cable temperature, x2 represents the conductor temperature, x3 represents the current, x4 represents the voltage, and x5 represents the loss value;

[0020] Step c: Map the input layer vector to the hidden layer, which has m neurons.

[0021] Step d: Get the output layer neuron vector o = (o1, o2, o3, o4) T , where o1 represents the temperature status value of the submarine cable, o2 represents the temperature status value of the conductor, o3 represents the current status value, and o4 represents the voltage status value.

[0022] Preferably, the method further comprises: normalizing the temperature state value of the submarine cable, the temperature state value of the conductor, the current state value, and the voltage state value to obtain the temperature state coefficient of the submarine cable, the temperature state coefficient of the conductor, the current state coefficient, and the voltage state coefficient;

[0023]

[0024] in, represents the state coefficient of the i-th item, o iIndicates output parameters: submarine cable temperature status value, conductor temperature status value, current status value and voltage data; imax and o imin They are the maximum and minimum values ​​set in the corresponding output parameters respectively.

[0025] Preferably, the method further comprises obtaining a fault evaluation index of the optoelectronic composite submarine cable:

[0026]

[0027] Where S represents the submarine cable fault evaluation index, S0 represents the submarine cable fault evaluation standard index, ξ represents the correction coefficient, represents the temperature state coefficient of the submarine cable, represents the conductor temperature state coefficient, represents the current state coefficient, represents the current state coefficient threshold, represents the voltage state coefficient, Indicates the voltage state coefficient threshold.

[0028] Preferably, step four includes:

[0029] When S≥90, the submarine cable is in normal use and has no faults;

[0030] When 70≤S<90, the submarine cable is in a sub-safe state, the monitoring frequency is doubled, and it continues to be used;

[0031] At that time, S was less than 70, and the submarine cable was in a faulty state, so the submarine cable was immediately inspected and repaired.

[0032] Preferably, the number of neurons in the hidden layer is 5; the activation functions of the hidden layer and the output layer are both S-type functions.

[0033] The beneficial effects of the present invention are:

[0034] 1. The mechanical fault monitoring method for the optoelectronic composite submarine cable provided by the present invention obtains the voltage and current of the optoelectronic composite submarine cable to calculate the loss value, further obtains the temperature data of the optoelectronic composite submarine cable, and determines the fault evaluation index of the optoelectronic composite submarine cable based on the BP neural network. This realizes the early judgment and timely warning of safety hazards, and effectively ensures the safe and stable operation of the submarine cable.

[0035] 2. The mechanical fault monitoring method of the optoelectronic composite submarine cable provided by the present invention obtains the temperature data of the submarine cable based on finite element simulation analysis, thereby improving the accuracy of the monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1This is a flow chart of a method for monitoring mechanical faults of an optoelectronic composite submarine cable provided by the present invention.

[0037] Figure 2 The figure is a flow chart of obtaining temperature data based on finite element simulation analysis in one embodiment of the present invention.

[0038] Figure 3 This is a finite element mesh division diagram of an optoelectronic composite submarine cable in one embodiment of the present invention.

[0039] Figure 4 The present invention is a flowchart of determining a fault evaluation index of an optoelectronic composite submarine cable based on a BP neural network in one embodiment of the present invention. DETAILED DESCRIPTION

[0040] The following describes the implementation of the present invention using specific embodiments. Those skilled in the art will readily understand the other advantages and benefits of the present invention from the disclosure herein. Obviously, the embodiments described are only a portion of the present invention, not all of it. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are intended to fall within the scope of protection of the present invention.

[0041] It should be noted that in the description of the present invention, terms such as "center," "upper," "lower," "lateral," and "inner" indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings. These are used solely for ease of description and are not intended to indicate or imply that a device or component must have, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limitations on the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0042] Furthermore, it should be noted that, in the description of the present invention, unless otherwise expressly specified or limited, the terms "disposed," "installed," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; they may refer to mechanical connections; they may refer to direct connections, indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0043] like Figure 1 As shown, the mechanical fault monitoring method of the optoelectronic composite submarine cable includes:

[0044] S110: Obtain the voltage and current of the optical-electrical composite submarine cable, and calculate the loss value according to the voltage and current.

[0045] The loss value is calculated as follows:

[0046]

[0047] Among them, W lose Indicates the loss value, represents frequency, C represents capacitance, represents voltage, I represents the current carrying capacity of the submarine cable, G Σ represents the conductance value, θ represents the phase angle, and e represents the base of the natural logarithm.

[0048] S120. Acquire temperature data of the optoelectronic composite submarine cable.

[0049] S130. Determine a fault evaluation index of the optoelectronic composite submarine cable based on a BP neural network.

[0050] S140. Adjust the monitoring frequency according to the fault evaluation index and make a prejudgment of the safety status of the submarine cable.

[0051] By obtaining the voltage and current loss values ​​of the optoelectronic composite submarine cable, and further obtaining the temperature data of the optoelectronic composite submarine cable, the fault evaluation index of the optoelectronic composite submarine cable is determined based on the BP neural network, which enables early judgment and timely warning of safety hazards, and effectively ensures the safe and stable operation of the submarine cable.

[0052] Furthermore, temperature data is obtained based on finite element simulation analysis. When establishing a submarine cable temperature field model using the finite element method, it is necessary to first establish a geometric model of the submarine cable and define some physical performance parameters of the submarine cable material under thermal analysis (for steady-state analysis, only the thermal conductivity of the material needs to be known). Then, the physical model is divided into appropriate grids, and finally the entire model is solved.

[0053] Specific steps are as follows Figure 2 Shown, including:

[0054] S121. Use ANSYS's adaptive meshing technology to perform triangular meshing on the submarine cable.

[0055] After determining the material properties and the geometric model of the submarine cable, the adaptive meshing technology of ANSYS is used to mesh the entire physical model. For the temperature field of the submarine cable, the most important thing is to obtain the conductor temperature and the optical fiber temperature, so the meshes of these two parts are further refined. The meshing results are as follows: Figure 3 shown.

[0056] S122. Perform steady-state thermal analysis on submarine cables immersed in seawater.

[0057] The main focus here is on the steady-state thermal analysis of submarine cables immersed in seawater. Therefore, only the third type of boundary conditions need to be considered: given the temperature of the fluid in contact with the surface and the heat transfer coefficient of the convection heat transfer surface, the heat flux density is:

[0058]

[0059] Where h represents the convective heat transfer coefficient, T w Indicates the surface temperature, T f represents the fluid temperature, λ represents the simulation coefficient, and n represents the grid division number.

[0060] S123. Apply the loss value to the corresponding loss area to perform simulation and solve to obtain the submarine cable temperature and the conductor temperature.

[0061] In order to obtain the temperature field of the submarine cable under the steady state, it is necessary to apply a load to the submarine cable, calculate the loss value of the submarine cable, convert the loss value into a heat generation rate and apply it to the corresponding loss area to solve the temperature field.

[0062] Further, such as Figure 4 As shown in FIG, the fault evaluation index of the optoelectronic composite submarine cable determined based on the BP neural network includes:

[0063] Step S131: Establish a three-layer BP neural network.

[0064] The BP network is a multi-layered network characterized by forward signal propagation and reverse error propagation. During forward propagation, a signal is input, which is then processed by the hidden layers until it reaches the output layer. The state of the neurons in the next layer is influenced by the neurons in the previous layer. When the output differs from the predicted result, reverse propagation occurs. The network weights and thresholds are then continuously adjusted based on the prediction error. Through this forward propagation and reverse propagation, the network parameters are updated. This continuous cycle of forward and reverse propagation continuously updates the network parameters to ensure that the BP neural network's predictions remain close to the predicted output.

[0065] The BP neural network architecture adopted in the present invention consists of three layers. The first layer is the input layer, which has a total of n nodes, corresponding to the n detection signals representing the submarine cable; the second layer is the hidden layer, which has a total of m nodes, which is determined in an adaptive manner by the network training process; and the third layer is the output layer, which has a total of p nodes, which is determined by the response actually required to be output by the system.

[0066] The mathematical model of the network is:

[0067] Input vector: x=(x1,x2,…x n ) T ;

[0068] Middle layer vector: y=(y1,y2,…y m ) T ;

[0069] Output vector: o=(o1,o2,…o p ) T ;

[0070] In a specific embodiment, the number of input layer nodes is n=5, the number of output layer nodes is n=5, and the number of hidden layer nodes is p=4.

[0071] The five parameters of the input layer are represented as follows: x1 represents the cable temperature, x2 represents the conductor temperature, x3 represents the current, x4 represents the voltage, and x5 represents the loss value;

[0072] The four parameters of the output layer are represented as follows: o1 represents the temperature state value of the submarine cable, o2 represents the temperature state value of the conductor, o3 represents the current state value, and o4 represents the voltage state value.

[0073] S132: Perform BP neural network training.

[0074] After the BP neural network node model is established, the BP neural network training can be carried out. The training samples are obtained based on historical experience data, and the connection weights between the input node i and the hidden layer node j, and the connection weights between the hidden layer node j and the output layer node k are given.

[0075] (1) Training methods

[0076] Each subnetwork is trained separately. During training, a set of training samples is first provided, where each sample consists of an input sample and an ideal output pair. When all the actual outputs of the network are consistent with their ideal outputs, the training is completed. Otherwise, the ideal output of the network is made consistent with the actual output by correcting the weights.

[0077] (2) Training algorithm

[0078] The BP network uses the Backward Propagation algorithm for training, and its steps can be summarized as follows:

[0079] Step 1: Select a network with a reasonable structure and set the initial values ​​of all node thresholds and connection weights.

[0080] Step 2: Calculate the following for each input sample:

[0081] (a) Forward calculation: for the j unit in layer l

[0082]

[0083] Where, represents the weighted sum of information of j units in layer l during the nth calculation, represents the connection weight between unit j in layer l and unit i in the previous layer (i.e., layer l-1), is the working signal sent by unit i in the previous layer (i.e., layer l-1, with nl-1 nodes); when i=0, let is the threshold of unit j in layer l.

[0084] If the activation function of unit j is the sigmoid function, then

[0085]

[0086] and

[0087]

[0088] If neuron j belongs to the first hidden layer (l=1), then

[0089] If neuron j belongs to the output layer, then and e j (n) = x j (n)-o j (n);

[0090] (b) Reverse calculation error:

[0091] For output units;

[0092]

[0093] For hidden units

[0094]

[0095] (c) Modified weights:

[0096] η is the learning rate.

[0097] Step 3: Input new samples or new cycle samples until the network converges. During training, the input order of samples in each cycle should be re-randomized.

[0098] The BP algorithm uses the gradient descent method to find the extreme value of nonlinear functions, which has the problems of falling into local minimum and slow convergence. A more effective algorithm is the Levenberg-Marquardt optimization algorithm, which shortens the network learning time and can effectively prevent the network from falling into local minimum. Its weight adjustment rate is

[0099] Δω=(J T J+μX) -1 J T β;

[0100] Where J represents the Jacobian matrix of the error differential with respect to the weights, X represents the input vector, β represents the error vector, and the variable μ is an adaptively adjusted scalar that determines whether learning is done according to Newton's method or the gradient method.

[0101] During system design, the system model is a pre-initialized network. Weights need to be learned and adjusted based on data samples acquired during use. This is why the system's self-learning function is designed to address this issue. Given a specified number of learning samples, the system can self-learn to continuously improve network performance.

[0102] S133. Normalize the submarine cable temperature state value, the conductor temperature state value, the current state value, and the voltage state value to obtain a submarine cable temperature state coefficient, a conductor temperature state coefficient, a current state coefficient, and a voltage state coefficient;

[0103]

[0104] in, represents the state coefficient of the i-th item, o i Indicates output parameters: submarine cable temperature status value, conductor temperature status value, current status value and voltage data; imax and o imin They are the maximum and minimum values ​​set in the corresponding output parameters respectively.

[0105] S134. Obtain a fault evaluation index of the optoelectronic composite submarine cable.

[0106]

[0107] Where S represents the submarine cable fault evaluation index, S0 represents the submarine cable fault evaluation standard index, ξ represents the correction coefficient, represents the temperature state coefficient of the submarine cable, represents the conductor temperature state coefficient, represents the current state coefficient, represents the current state coefficient threshold, represents the voltage state coefficient, Indicates the voltage state coefficient threshold.

[0108] Furthermore, step S140 includes:

[0109] When S≥90, the submarine cable is in normal use and has no faults;

[0110] When 70≤S<90, the submarine cable is in a sub-safe state, the monitoring frequency is doubled, and it continues to be used;

[0111] At that time, S was less than 70, and the submarine cable was in a faulty state, so the submarine cable was immediately inspected and repaired.

[0112] Preferably, the number of neurons in the hidden layer is 5; the activation functions of the hidden layer and the output layer are both S-type functions.

[0113] The present invention obtains the voltage and current loss values ​​of the optoelectronic composite submarine cable, further obtains the temperature data of the optoelectronic composite submarine cable, and determines the fault evaluation index of the optoelectronic composite submarine cable based on the BP neural network, thereby achieving early judgment and timely warning of safety hazards, and effectively ensuring the safe and stable operation of the submarine cable.

[0114] The above contents are merely embodiments of the present invention. Common knowledge such as the specific structures and characteristics of the solutions is not described in detail herein. It is readily understood by those skilled in the art that the scope of protection of the present invention is not limited to these specific embodiments. Numerous variations and improvements may be made without departing from the present invention, and these variations and improvements should also be considered within the scope of protection of the present invention. These variations and improvements will not affect the effectiveness of the implementation of the present invention or the practicality of the patent.

Claims

1. A method for monitoring mechanical faults of an optoelectronic composite submarine cable, characterized in that: include: Step 1, obtaining the voltage and current of the optoelectronic composite submarine cable, and calculating the loss value according to the voltage and current; The calculation formula of the loss value is: Among them, W lose Indicates the loss value, represents frequency, C represents capacitance, represents voltage, I represents the current carrying capacity of the submarine cable, G Σ represents the conductance value, θ represents the phase angle, and e represents the base of the natural logarithm; Step 2: obtaining temperature data of the optoelectronic composite submarine cable; The temperature data is obtained based on finite element simulation analysis, which specifically includes the following steps: Use ANSYS's adaptive meshing technology to perform triangular meshing on submarine cables; Steady-state thermal analysis of a submarine cable immersed in seawater: Where h represents the convective heat transfer coefficient, T w Indicates the surface temperature, T f represents the fluid temperature, λ represents the simulation coefficient, and n represents the grid division number; Applying the loss value to the corresponding loss area to perform simulation and solve to obtain the temperature of the submarine cable and the conductor; Step 3: determining the fault evaluation index of the optoelectronic composite submarine cable based on the BP neural network; Among them, step three specifically includes: Step a, establish a three-layer BP neural network; Step b: Determine the input layer neuron vector x = (x1, x2, x3, x4, x5) T , where x1 represents the cable temperature, x2 represents the conductor temperature, x3 represents the current, x4 represents the voltage, and x5 represents the loss value; Step c, mapping the input layer vector to the hidden layer, where the number of neurons in the hidden layer is m; Step d: Get the output layer neuron vector o = (o1, o2, o3, o4) T , where o1 represents the temperature state value of the submarine cable, o2 represents the temperature state value of the conductor, o3 represents the current state value, and o4 represents the voltage state value; The method further includes: normalizing the temperature state value of the submarine cable, the temperature state value of the conductor, the current state value, and the voltage state value to obtain a temperature state coefficient of the submarine cable, a temperature state coefficient of the conductor, a current state coefficient, and a voltage state coefficient; in, represents the state coefficient of the i-th term, o i Indicates output parameters: submarine cable temperature status value, conductor temperature status value, current status value and voltage data; imax and imin They are the maximum and minimum values ​​set in the corresponding output parameters respectively; It also includes obtaining a fault evaluation index of the optoelectronic composite submarine cable: Where S represents the submarine cable fault evaluation index, S0 represents the submarine cable fault evaluation standard index, ξ represents the correction coefficient, represents the temperature state coefficient of the submarine cable, represents the conductor temperature state coefficient, represents the current state coefficient, represents the current state coefficient threshold, represents the voltage state coefficient, Indicates the voltage state coefficient threshold; Step 4: adjusting the monitoring frequency according to the fault evaluation index, and predicting the safety status of the submarine cable; The fourth step comprises: When S ≥ 90, the submarine cable is in normal use without any faults; When 70≤S<90, the submarine cable is in a sub-safe state, the monitoring frequency is doubled, and it continues to be used; At that time, S<70, the submarine cable is in a faulty state, and the submarine cable is immediately repaired.

2. The method for monitoring mechanical faults of an optoelectronic composite submarine cable according to claim 1, characterized in that: The number of neurons in the hidden layer is 4; the activation functions of the hidden layer and the output layer are both S-type functions.

Images